Advances in sensor technology, distributed control
systems and computer technology are dramatically increasing the amount
of data that can be collected on the operation of industrial plants. However,
the sheer volume of these inputs may make it extremely difficult for human
operators to react appropriately. The aim of the FP5 CHEM project is to
develop decision support system software that will integrate the use of
specialised process control and simulation 'toolboxes' to provide usable
on-line information for efficient, problem-free management of refining,
chemical and petrochemical processes.
Around the world, unscheduled plant shutdowns and
other abnormal situations are causing economic losses amounting to billions
of euros each year, plus untold damage caused by serious accidents.
Today's control systems include more and more monitoring
equipment - but this is mainly to provide more data to operators who still
have to make the important and often complex decisions in a very limited
time. Incidents such as Three Mile Island, Bhopal, and Chernobyl are chilling
examples of faults that turned into disasters, partly due to improper
actions on the part of operators probably faced with an information overload.
The objective of the GROWTH CHEM (1)
project is to eliminate such risks and facilitate the achievement
of optimal operating states by greatly simplifying decision making
in refining, chemical and petrochemical operations.
While much work has already been carried out
in process trend analysis, fault diagnosis and decision support,
the resultant tools typically function as separate entities. It
thus remains difficult for the industrial designer to build fully
integrated supervision systems for new plants. CHEM seeks to solve
this problem by providing a flexible decision support system (DSS)
capable of digesting and interpreting the data delivered by a wide
range of disparate measurement, modelling and analysis software
This three-year GROWTH project began in April
2001, with an eight-nation consortium including both EU and applicant
country partners and led by the Institut Français du Pétrole.
Participants include five industrial companies and five academic
institutions, with four industrial assistant contractors. The scope
of CHEM is further extended by its incorporation into the international
IMS (Intelligent Manufacturing Systems) programme, adding more partners
from the USA and Japan.
The DSS to be developed will allow synergistic
combination of advanced tools developed by the team members, who
bring complementary expertise in various aspects of process supervision.
The University des Sciences et Technologies de Lille (USTL) and
THALES Airborne Systems, together with the Warsaw University of
Technology (WUT), are specialists in fault diagnosis. The Universitat
de Girona (UdG) leads in process trend analysis, and the Universitat
Politecnica de Catalunya (UPC) in reactive scheduling. Lund University
has done important work in decision support structures, while the
Technical Research Centre of Finland is a major actor in safety
and risk analysis. Software vendor Computas is deeply involved in
statistical process control. Gensym is a specialist in on-line integrated
software products that model, simulate and manage critical operations
across a broad range of industries.
End users will contribute the necessary experience
in handling real industrial situations and developing complete supervision
CHEM: Advanced decision support system for chemical/petrochemical
manufacturing processes (G1RD-CT-2001-00466)
A demonstrator package will be built around
software widely used in industry, with a modular design that will
readily allow new supervision applications to be incorporated as
To prove the concept, industrial and academic
partners will provide access to pilot and full-scale plants running
processes ranging from Fluid Catalytic Cracking (IFP) to coal gasification
(UPC), steam generation (USTL), papermaking (UPM-Kymmene), benzole
recovery (Corus) and hydroxylamine sulphate manufacturing (ZAP -
a subcontractor to WUT). This diversity will enable the broad applicability
of the approach to be verified under actual working conditions.
Anticipated exploitable results of the project
- Methodology, whereby advanced process supervision
and optimisation applications can be designed and built using
the common CHEM architectural framework and individual toolboxes.
The essentials of the methodology will be made public, in order
to ensure as wide as possible industrial take-up.
- Toolboxes - individual software modules that
can be 'plugged into' the common framework.
- Applications - complete software solutions
built with relevant toolboxes and dedicated to specific processes.
Reducing accidents and pollution
By preventing or managing abnormal events, CHEM will reduce accidents
and pollution - thus playing a major role in safety and environment
protection. In addition, it will yield valuable new knowledge about
process parameters and critical conditions. Users will benefit from
the ability of integrated supervision to improve the efficiency
of their operations, and thus reduce costs. It will also enhance
working conditions by easing operators' tasks and removing potential
sources of stress.
Institut Français du Pétrole
Avenue Napoleon Bonaparte 232
F-92500 Rueil Malmaison, France
Tel: +33 1 47 52 70 17
Fax: +33 1 47 52 70 22
- Computas, Norway
- Corus UK
- Gensym, The Netherlands
- Lund University of Technology, Sweden
- Metso, Finland
- Nitrogen Works 'Pulawy', Poland
- Oy Keskuslaboratorio - Centrallaboratorium,
- Technical Research Centre of Finland, Finland
- THALES Airborne Systems, France
- Universitat Politecnica de Catalunya, Spain
- Universitat de Girona, Spain
- Universite des Sciences et Technologies de
- UPM-Kymmene, Finland
- Warsaw University of Technology, Poland